PERGAMON Energy Conversion & Management 40 (1999) 175±191

Transient performance analysis of cylindrical parabolic

concentrating collectors and comparison with experimental
results
N. Eskin
Faculty of Mechanical Engineering, ÇIstanbul Technical University, GuÈmuÈs° suyu, ÇIstanbul, Turkey
Received 19 February 1998

Abstract

This paper presents the unsteady, one-dimensional performance analysis of cylindrical parabolic
concentrating collectors through the ®rst and second laws of thermodynamics. These laws, together, help
to de®ne the optimum system that satis®es the imposed thermal and economical constraints and
minimizes exergy loss. The analysis considers the thermal masses of the absorber pipe, pyrex envelope
and the working ¯uid of the system individually, in an unsteady state. The instantaneous exergetic and
energy eciencies for the cylindrical parabolic collectors are compared for daily insolation at di€erent
¯ow rates. The values measured using the MKE Institute Collector Test Facility are in good agreement
with the predicted simulation results. The simulation results indicate that the exergetic eciency is
highly dependent on the mass ¯ow rates and the arrangement of the cylindrical parabolic collector
array. The optimum ¯ow rates which provide minimum exergy loss were obtained from the simulation
model. # 1998 Elsevier Science Ltd. All rights reserved.

Keywords: Cylindrical parabolic collectors; Second law analysis; Exergetic eciency

Nomenclature

A = Area (m2)
c = Speci®c heat (J/kg K)
d = Diameter (m)
E = Energy (J)
f = Friction factor
g = Acceleration of gravity (m/s2)
h = Convective heat-transfer coecient (W/m2 K)
I = Solar insolation incident on collector (W/m2)

0196-8904/99/$ - see front matter # 1998 Elsevier Science Ltd. All rights reserved.
PII: S 0 1 9 6 - 8 9 0 4 ( 9 8 ) 0 0 0 3 5 - 1
176 N. Eskin / Energy Conversion & Management 40 (1999) 175±191

k = Thermal conductivity, (W/m-K)

L = Absorber length (m)
mÇ = Mass ¯ow rate (kg/s)
m = Speci®c ¯ow rate (kg/s-m2)
P = Periphery (m)
Dp = Pressure drop (Pa)
q = Heat ¯ux (W/m2)
T = Temperature (K)
t = Time (s)
U = Overall heat transfer coecient (W/m2 K)
V = Velocity (m/s)
w = Re¯ector aperture (m)

Greek
a = Absorptance (ÿ)
e = Emissivity (ÿ)
g = Interception factor (ÿ)
Z = Collector eciency (ÿ)
k = Dilution factor (ÿ)
r = Density (kg/m3)
s = Stefan±Boltzman constant (W/m2 K4)
ta = Absorptance-transmittance product (ÿ)
X = Exergy (J)
C = Available energy ¯ux (W/m2)

Subscripts
a = Absorber pipe
ap = Absorber pipe to pyrex envelope
be=Beam
c = Collector, convective
d = Di€use
e = E€ective
i = Inside, inlet
p = Pyrex envelope
pa = Pyrex envelope to absorber
po = Pyrex envelope to surrounding
pr = Pressure
R = Reference state
r = Re¯ector, radiative
s = Sun
o = Outside, outlet, surrounding
w = Water
 N. Eskin / Energy Conversion & Management 40 (1999) 175±191 177

1. Introduction

In thermal applications of solar energy, it is often desired to obtain a high ¯uid temperature,
since the eciencies of power generation systems and refrigeration systems increase with
temperature. Solar energy collection at temperatures around 2008C is highly desirable,
especially for process heat generation systems. Among the geometries of cylindrical parabolic
collectors which have been developed so far, the cylindrical parabolic concentrating collectors
are the most economical devices.
The solar radiation reaching the earth's surface in any location varies considerably with
atmospheric conditions, hour of the day and season of the year. Therefore, solar heat
collectors always operate in an unsteady state. Also, the conversion of solar radiation into
thermal energy and the transfer of this energy to the working ¯uid are all entropy generating
(irreversible) processes, together with the associated thermal losses to the ambient. When these
processes occur with minimum irreversibility, a high performance level can be expected. This
concept has found increasing interest in recent years [12±14]. In the literature, there are studies
about steady state exergy analysis of various types of solar collectors [2, 5, 6] and unsteady
state exergy analysis of ¯at plate collectors [10].
Transient exergy analysis of cylindrical parabolic solar collectors is particularly useful in
their design. Exergy analysis provides the basis for choosing the operating parameters of the
solar collectors. The requirements for greater conversion eciency and the introduction of new
devices have led to the need for improved methods of prediction of design parameters.
Performance analysis of cylindrical parabolic collectors through exergy analysis has led the
designer to improve the design parameters.
In this article, the performances of cylindrical parabolic collectors for process heat
generation have been analysed through the ®rst and second laws of thermodynamics for the
collectors produced by the MKE (Mechanical and Chemical Developments Institute). The
analysis considers the thermal masses of the pyrex envelope, absorber pipe and the working
¯uid of the system individually. The variation of the pyrex envelope temperature, absorber
temperature and the working ¯uid temperature will be evaluated along the collector length at
every hour during a day.
The e€ect of mass ¯ow rates and types of collector arrangements on the instantaneous
energetic and exergetic eciencies will be examined. Simulation results will be compared with
experimental data obtained at di€erent ¯ow rates and at di€erent inlet water temperatures. The
variation of optimum mass ¯ow rates with respect to solar time will be obtained for a given
insolation by minimizing the instantaneous exergy loss in the collector. The e€ect of collector
arrangement on the collector eciency will be obtained.

2. Analysis

A schematic drawing of the solar collector system under consideration is shown in Fig. 1.
This collector consists of a cylindrical parabolic re¯ector, a pyrex envelope and an absorber
pipe. The absorber pipe, which is coated with a selective surface coating, is centered along the
re¯ector's focal line.
178 N. Eskin / Energy Conversion & Management 40 (1999) 175±191

Fig. 1. A schematic of the cylindrical parabolic collector.

1. In this analysis, the temperatures of the absorber pipe and pyrex envelope are assumed to
be circumferentially uniform, as stated in [8] for the working ¯uid which is completely in the
liquid phase.
2. In the simulations, the temperature gradients through the thickness of the walls of the
absorber and the envelope are assumed to be negligible.
3. The mass ¯ow rate of the working ¯uid is considered to be constant and the heat transfer
from the absorber pipe to the working ¯uid is of convective nature.

Under the assumptions stated above and with the di€erential control volumes around the
absorber pipe and the pyrex envelope as given in Fig. 2(a) and 2(b), respectively, the energy
balances for the envelope, the absorber pipe and the working ¯uid can be expressed as follows.

2.1. Absorber pipe

@2 Ta @Ta
Aa ka 2
‡ ta†e gr Iw ÿ pdao Uap Ta ÿ Tp † ÿ pdai hi Ta ÿ Tw † ˆ ra ca Aa 1†
@x @t
where Uap is the overall heat transfer coecient between the absorber pipe and the pyrex
envelope and Ta, Tp, Tw the absorber, pyrex envelope and water temperatures, respectively. Uap
 N. Eskin / Energy Conversion & Management 40 (1999) 175±191 179

Fig. 2. Energy balances. (a) Absorber pipe energy balance, (b) pyrex envelope energy balance.

is the combination of radiative and convective heat transfer coecients Uap = hr,ap + hc,ap, and
the radiative component is
s Tp ‡ Ta † T2p ‡ T2a †
hr;ap ˆ : 2†
1=Ep ‡ 1=Ea ÿ 1
The e€ective convective coecient is obtained from correlations given in [7].
The absorptance-transmittance product (ta)e varies with the solar incidence angle and, so,
depends on the time of day in a manner given in [3].

2.2. Pyrex envelope

An envelope is used in order to decrease the convective heat transfer from the absorber pipe
to the surroundings.
@2 Tp @Tp
Ap kp 2
‡ ap gr Iw ÿ pdpi Upa Ta ÿ Tp † ÿ pdpo Upo Tp ÿ T0 † ˆ rp cp Ap 3†
@x @t
180 N. Eskin / Energy Conversion & Management 40 (1999) 175±191

where Ap = p/4(d 2po ÿd 2pi) is the cross sectional area of the annulus. Here, Upo is the overall
heat transfer coecient between the pyrex envelope and its surroundings, T0 is the
surroundings temperature and a p is the absorption coecient of the pyrex envelope.

2.3. Collector ¯uid

A di€erential formulation to include terms due to thermal capacity, axial heat conduction
and heat transfer from the absorber tube to the working ¯uid leads to the following equation;
@Tw @Tw
rw cw Aw ‡ m_ w cw ˆ pdai hi Ta ÿ Tw † 4†
@t @x
where Aw is the ¯ow cross-sectional area. For the initial conditions of the problem, it's
assumed that the entire system is in thermal equilibrium with the ambient at sunrise.

3. Solar radiation availability and exergy analysis of cylindrical parabolic collector

The expression of the solar radiation availability and the radiation temperature are well
documented in the literature [1, 4, 9]. The origins and the correctness of the expressions for the
apparent sun temperature have been widely discussed in these papers. The expression for the
availability of solar radiation with beam and di€use components has been used in this paper as
given in [10].
   
4T0 4T0
c ˆ Ibe 1 ÿ ‡ Id 1 ÿ 5†
3Ts 3T s
where
Ts =T s ˆ 0:9562 ‡ 0:2777 ln 1=k† ‡ 0:0511 k: 6†
The expression for Ts/T*s relating the e€ective temperature of di€use radiation to the solar
temperature, is obtained from [9] with k, the dilution factor of di€use radiation, less than 0.1.
In order to account for the e€ects of the thermal capacities of the various components of the
solar collector, the collector exergetic eciency has been examined in this paper. During
operation, while the ¯uid temperature increases from an inlet value of Ti, to an outlet value of
To, changes occur in the local absorber and envelope temperatures. Hence, at any given time,
the energy and exergy outputs consist of the quantity extracted by the collector ¯uid and the
quantity stored, or released, in the absorber and envelope. The expression for the rate of
energy gain can, therefore, be written as;
Z Z
_ @ L @ L
E ˆ m_ w cw Tw;o ÿ Tw;i † ‡ ra ca Aa Ta dx ‡ rp cp Ap Tp dx 7†
@t 0 @t 0
and, in a similar manner, the rate of exergy gain can be expressed as follows:
 N. Eskin / Energy Conversion & Management 40 (1999) 175±191 181
Z    Z   
d L Ta d L Tp
w_ ˆ ra ca Aa Ta ÿTR † ÿTR ln dx ‡ rp cp Ap Tp ÿTR † ÿTR ln dx
dt 0 TR dt 0 TR
   Z   
To d L Tw m_ w Dp
‡cw m_ w To ÿ Ti ÿ TR ln ‡ rw cw Aw Tw ÿ TR ÿ TR ln dx ‡ 8†
Ti dt 0 TR rw
Dp is the pressure drop between the inlet and outlet points of the collector and is given by [5]
as:
V2w
Dp ˆ frw L : 9†
2dai
Here, f is the friction coecient given by [11].
The energetic eciency, de®ned as the ratio of energy output to energy input, becomes
_ cI
ZI ˆ E=A 10†
while the second law eciency, de®ned as the ratio of the exergy gained by the collector to the
availability of the solar radiation, becomes
w_
ZII ˆ 11†
Ac ‰Ibe 1 ÿ 4T0 =3Ts † ‡ Id 1 ÿ 4T0 =3T s †Š
where wÇ is the sum of the exergy gained by the absorber, the pyrex envelope and the working
¯uid and the exergy lost by the water pressure, i.e.
w_ ˆ w_ a‡p ÿ w_ w ‡ w_ pr;w : 12†

4. Test facility for solar collector

The schematic layout of the test unit used in the experiments is shown in Fig. 3(a). The test
unit consists of east±west aligned cylindrical collectors (44 m2); a pressurized tank and a
circulation pump. A view of the cylindrical parabolic collector is shown in Fig. 3(b).
The cylindrical parabolic collectors in the test unit consist of cylindrical parabolic re¯ectors
with a pyrex envelope and a 0.15% steel absorber pipe connected along the length of the
system. The pyrex envelope and the absorber pipe are coaxially placed at the focal line of the
re¯ectors. The mirrors that are used as re¯ectors are parabolic in shape and plated with silver
oxide. They are coated with a speci®c surface coating to protect them from environmental
e€ects. The collectors are provided with a driving system to track the sun and maintain
focusing of the solar radiation on the pyrex envelope and the absorber.
The heat transfer medium used in the experiments is pressurized water. This medium ¯ows
in the state of liquid phase in the absorber pipe. The mass ¯ow rate was kept constant while
obtaining each set of data.
The measurements were taken with 26 thermoresistant sensors for the temperatures, and ¯ow
meters in each loop were used for the volume ¯ow rate. The beam radiation was measured by
a pyrheliometer. The pyranometer mounted on the platform measured the global insolation,
182 N. Eskin / Energy Conversion & Management 40 (1999) 175±191

Fig. 3. The cylindrical parabolic collector test unit (a) The schematic layout, (b) a view of the collectors in MKE
Institute.

while further radiometers recorded the global radiation on the horizontal surface and the
di€use radiation.
Ambient temperature and wind velocity and direction were also measured during the
experiments. The expected errors are 20.75 % for the temperature measurements and 22% for
the volume ¯ow measurements.
To test and validate the formulation presented in this paper, the daily performance of the
cylindrical parabolic collector was simulated for an Ankara day using the set of insolation data
 N. Eskin / Energy Conversion & Management 40 (1999) 175±191 183

Table 1. Measured global and di€use insolation and the calculated available insolation values (k = 0.001;
TR = 293 K)

Solar time Global insolation (W/m2) Di€use insolation (W/m2) Available insolation (W/m2)

6 123 15. 113.1

7 244 30 224.2
8 471 58 433.2
9 666 86 612.5
10 815 104 750.3
11 910 117 836.9
12 942 119 866.6
13 918 111 843.9
14 810 97 754.
15 660 83 606.5
16 474 63 433.2
17 243 33 224.2
18 121 15 111.2
19 11 Ð 10

gathered on 18 July, 1990. The set of data is for a clear day with a global insolation, from
sunrise to sunset, of 26.7 MJ/m2 (measured), where the availability of 24.5 MJ/m2, for
insolation was calculated from measured values by using Eq. (5). On this day, the di€use
radiation is only a small fraction of the global radiation. The measured insolation value varies
monotonically with time and is near sinusoidal with a maximum value around solar noon. The
measured values of the hourly global and di€use insolation and the calculated available
insolation values are given in Table 1.
The same insolation values were used in the simulations. The key data of the installed
cylindrical parabolic collectors which were used in the experiments are given in Table 2.
The cylindrical parabolic collector has been extensively tested at the MKE Institute and is
chosen because its performance record is well established. This allows the use of their test data
to verify the prediction results given in this paper. These data were also used as the computer
input to yield the simulation results.

Table 2. Collector speci®cations

Manufacturer MKE/Turkey
Concentrating ratio 48
Area of one module 2 m  11 m
Absorber pipe Steel
Diameter 0.042 m/0.037 m
Absorptance/re¯ectance 0.85/0.15
Envelope Pyrex
Diameter 0.066 m/0.0623 m
Transmittance/absorptance 0.90/0.05
Re¯ector re¯ectance 0.90
184 N. Eskin / Energy Conversion & Management 40 (1999) 175±191

Fig. 4. Collector water outlet temperature, energetic and exergetic values along a solar day.
 N. Eskin / Energy Conversion & Management 40 (1999) 175±191 185

5. Results and discussion

The simulation of the cylindrical parabolic collector enables us to vary the parameters of
system design or operation. Di€erent operation parameters can be examined and compared
under the same weather conditions without expensive outdoor tests. In this way cylindrical
parabolic collectors which are used in process heat plants can be designed with a higher
eciency to reduce the unit cost of the energy obtained.
The developed simulation program was used in obtaining the energetic and exergetic (second
law) eciencies of cylindrical parabolic collectors which operate in heat process plants using
hot water. In the simulations, an unsteady state model of the collector system is used.
Performance results for speci®c ¯ow rates (¯ow rate per collector area) between 0.002 kg/s
m2 and 0.026 kg/s m2 are presented in Fig. 4. The test and predicted results shown in Fig. 4
are obtained for 208C inlet water temperature. As shown in Fig. 4, the ¯uid ¯ow rates and
insolation have a dominant in¯uence on the exergetic performance of the solar collector. The
®rst law eciencies are directly and the second law eciencies, and temperatures, are indirectly
proportional to the ¯ow rates. The exergetic eciency curves tend to follow the insolation
pattern, while the ®rst law eciency curves vary between 53±56%, independent of the
insolation pattern.
In order to observe the e€ect of higher water temperatures on the exergetic eciencies of the
collector, a storage tank (V = 0.225 m3) was used to provide water at increasing temperatures,
thus resulting in higher values for the water temperatures of the test unit (Fig. 3(a)). The
experimental values were obtained for series connected cylindrical parabolic collectors with a
total area of 44 m2 for two di€erent mass ¯ow rates along the receiver (absorber tube and

Fig. 5. Comparison of the experimental and theoretically calculated values for two di€erent speci®c ¯ow rates
m col = 0.01 kg/s m2 and m col = 0.0019 kg/s m2.
186 N. Eskin / Energy Conversion & Management 40 (1999) 175±191

Fig. 6. Collector water temperature, absorber temperature and envelope temperatures along the absorber tube
length in a module at solar noon.

envelope) length. During the experiments, the load mass ¯ow rate was taken as
m load = 0.3 m col. The results were compared with experimentally determined temperatures
(Figs. 5 and 6).
Figure 6 shows the comparison of the experimental and simulation results along the receiver
at solar noon. As it is shown in Figs. 5 and 6, the maximum di€erence between the measured
and calculated temperature values is about 58C (3%) for the water temperatures. This
di€erence is due to the uncertainties in the temperature measurements 20.75%, ¯ow
measurements 22% and speci®c heat capacity 26%. Simulation results agree with the
experimental results and verify the formulation of the problem.
Figures 7(a), (b) and (c) illustrate the simulation results of the dominant in¯uence of ¯uid
¯ow rates between 0.0005±0.026 kg/s m2 on the performance of cylindrical parabolic collectors.
As a result, it seems that, within a range, a continued increase of ¯ow rate would produce a
continued increase in the exergetic eciency, but this phenomenon changes as the increase in
the absorber pipe temperature gives rise to increased exergy losses from the collector receiver
system. After a speci®c mass ¯ow rate of 0.01 kg/s m2 decrease in exergetic eciency was
obtained from the simulation results.
Exergetic eciency is the measure of entropy generation or the exergy loss of the system.
The decrease in exergetic eciency is due to the amount of entropy generation during the
processes that occur in the cylindrical parabolic collectors.
N. Eskin / Energy Conversion & Management 40 (1999) 175±191 187

Fig. 7. Collector performance values as a function of solar time.

188 N. Eskin / Energy Conversion & Management 40 (1999) 175±191

The entropy generation due to the conversion of solar radiation to thermal energy is the
most substantial exergy loss in most solar collectors. This exergy loss is comparatively smaller
in cylindrical parabolic collectors than in the other collector systems due to the high
performance in focusing and tracking systems [15]. This loss can also be lowered by increasing
the absorber tube temperature through improvements in the optical properties of the receiver
system, such as increased absorptance of the absorber tube and increased transmittance of the
envelope. The absorber tube temperature can be increased by decreasing the mass ¯ow rates.
The thermal masses of the absorber tube and the envelope generate entropy which act as
storage or accumulation of exergy as the receiver system temperature increases. With the
temperature decrease, it acts as the releasor of exergy. Simulation results indicate that the
storage of exergy is important, especially after sunrise when the component temperatures start
increasing. During this period, most of the energy collected from the sun is used to heat the
components, while a small amount of energy is transferred to the collector ¯uid. This exergy
loss can be controlled by increasing the mass ¯ow rates (Figs. 6 and 7).
A third important parameter in exergy loss is the thermal energy losses by both convective and
radiative heat transfer from the absorber pipe and the envelope. This also can be controlled by
reducing the temperature of the system components. Thermal energy losses with radiative heat
transfer can also be reduced by the use of absorber tubes with lower surface emissivity.
The other processes in the system responsible for entropy generation are:
ÐThe thermal conduction in the absorber pipe and its envelope. This term is small due to
the small temperature gradients through the pipes' thicknesses. Test results have shown that
the temperature di€erence through the cross section of the walls of the absorber pipe is
negligible due to its high thermal conductivity.
ÐThe heat transfer from the absorber pipe to the working ¯uid at ®nite temperature
di€erence.
ÐT#he friction between the collector ¯uid and the pipe. The calculations showed that the
exergy loss due to friction between the collector ¯uid and the pipe is important, especially after
the speci®c ¯ow rate of 0.0038 kg/s m2 for the parameters used in the simulations. This loss can
be controlled for the tested MKE collector by keeping the speci®c ¯ow rates below this value.
The exergetic eciency is an indication of the entropy generation or exergy destruction by
the system. For optimum operation of the cylindrical parabolic collector, the sum of the
entropy generated or the exergy losses must be minimized. The total exergy output of a
cylindrical parabolic collector can be expressed as the following integral for an interval of time.
Zt
wˆ w_ dt: 13†
0

Among the independent parameters which a€ect the maximization of this integral are the
collector ¯uid ¯ow rate, insolation, collector geometry and material properties. One of the
operation parameters which a€ects entropy generation substantially is the mass ¯ow rate. By
integrating among di€erent time intervals, the mass ¯ow rates which optimize this exergy
output integral are determined. A numerical approach was used, varying the ¯ow rate until the
system exergy output is a maximum within the control period. The results are presented in
Fig. 8. The results show that the optimum ¯ow rate depends on the control period. The
 N. Eskin / Energy Conversion & Management 40 (1999) 175±191 189

Fig. 8. Optimum ¯ow rates for two di€erent ¯ow control period.

optimal speci®c ¯ow rate is nearly constant (0.0023 kg/s m2) for the entire insolation period
from sunrise to sunset for the insolation data presented earlier.
In solar±thermal conversion systems, a number of collectors must be connected to provide
sucient area to capture energy for the designed end use. For that reason, a parallel, series, or
series-parallel con®guration is generally used in large arrays of cylindrical parabolic collectors.
The e€ects of collector arrangement on the energetic and exergetic eciencies are represented
in Fig. 9. As it is observed from the ®gure, collectors connected in parallel, and series, or
series-parallel almost have the same ®rst law eciencies.
Exergetic eciencies are also compared for the series, parallel and series-parallel connected
cylindrical parabolic collectors for the same water outlet temperatures obtained during the day,
and for the same speci®c ¯ow rate. From the simulation results presented in Fig. 9, series
connected collectors have higher exergetic eencies than the parallel connected collectors;
while series and series-parallel connected collectors have the same exergetic eciencies at every
solar time during the day.

6. Conclusions

In this paper, a more complete analysis of the cylindrical parabolic concentrating collector
subjected to insolation which varies by time have been analyzed. Absorber pipe, envelope and
collector ¯uid were considered separately and analytical equations were developed.
190 N. Eskin / Energy Conversion & Management 40 (1999) 175±191

Fig. 9. E€ect of the collector arrangement on collector performance.

 N. Eskin / Energy Conversion & Management 40 (1999) 175±191 191

The performance of cylindrical parabolic collectors was analyzed from the standpoint of
exergy.
The values obtained from the experiment are found to be in agreement with the developed
simulation program results. Using measured insolation values, the energetic and exergetic
eciencies of the collectors were determined. The exergetic eciency was found to be strongly
dependent on the collector working ¯uid ¯ow rates and insolation. By minimizing the exergy
loss among di€erent time intervals, the optimum mass ¯ow rates for 1/2 h and 1 h control
periods were obtained for series connected cylindrical parabolic collectors.
The arrangement of the collectors in series and parallel arrays was also analysed. The e€ect
of the arrangement of the collectors on the energetic eciency is negligible. Series or series-
parallel connected collector arrangements have higher exergetic eciencies than parallel
connected collectors at every solar time during the day.

References

[1] Bejan A. Uni®cation of three di€erent theories concerning the ideal conversion of enclosed radiation. Trans. of
the ASME 1987;109:46.
[2] Chalghoum DE, Bejan A. Second-law analysis of solar collectors with energy storage capability. Trans. of the
ASME 1985;107:244.
[3] Due JA, Beckman WA, Solar Energy Thermal Processes, Wiley-Interscience Publication, 1974.
[4] Gribik JA, Osterle JF. The second law eciency of solar energy conversion. Trans. of the ASME 1987;106:16.
[5] Fujiwara M. Exergy analysis for the performance of solar collectors. Trans. ASME, J. Solar Energy Eng.
1983;105:163.
[6] Han KW, Lee C C. A second law concept of an equivalent solar collector. Energy 1991;16(5):859.
[7] Holland KGT, Unny TE, Raithby GD, Konicek L. Free convective heat transfer across inclined air layers.
Trans. ASME. J. Heat Transfer 1976;98:189.
[8] Heidemann W, Spindler K, Hahne E. Steady-state and transient temperature ®eld in the absorber tube of a
direct steam generating solar collector. Int. J. Heat Mass Transfer 1992;35(3):649.
[9] Landsberg PT, Tonge G. Thermodynamics of the conversion of diluted radiation. J. Phys. A: Math. Gen.
1979;12:551.
[10] Onyegegbu SO, Morhenne J. Transient multidimensional second law analysis of solar collectors subjected to
time-varying insolation with di€use components. Solar Energy 1993;50(1):85.
[11] Rohsenow WM, Hartnett, JP, Handbook of Heat Transfer, McGraw-Hill, New York, 1973, pp. 7±73.
[12] Said SAM, Zubair SM. On second-law eciency of solar collectors. Trans. of the ASME 1993;115:2.
[13] Suzuki A. A fundamental equation for exergy balance on solar collectors. Trans. of the ASME 1988;110:102.
[14] Suzuki A, Okamura H, Oshida I. Application of exergy concept to the analysis of optimum operating con-
ditions of solar heat collectors. J. Solar Energy Engineering 1987;109:337.
[15] Thomas A, GuÈven HM. Parabolic trough concentratorsÐDesign, construction and evaluation. Energy Conver.
Mgmt. 1993;34(5):401.